Effect of medical complications on the after-stroke rehabilitation outcome

Abstract

BACKGROUND: Post-stroke rehabilitation should begin immediately and continue while improvement is seen. After stroke, patients are at risk of developing medical complications, which can hinder optimal recovery.

OBJECTIVE: We aimed to determine the incidence of complications occurring during early inpatient rehabilitation and to assess their influence on patients’ functionality.

METHODS: For 1075 patients (585 men), we recorded neurological deficits, activities of everyday living, functionality before and after rehabilitation, and the incidence of complications. The main measure of functional outcome was the discharge Barthel Index score.

RESULTS: At least one complication was reported by 76.9% of patients, and 20% experienced three or more complications. The most common problems were: urinary tract infection (23.2%), depression (18.9%), falls (17.9%), unstable hypertension (17.6%), and shoulder pain (14.9%). Patients with a severe discharge disability were 2.5-fold more likely to experience complications than subjects with a mild disability.

CONCLUSIONS: Complications during the post stroke rehabilitation process had a significantly decreased chance for improvement. Knowledge of the most common type of complications, as well as their early prevention and treatment, may improve patient outcomes.

Keywords

Stroke complications neurorehabilitation

1 Introduction

Stroke is a leading cause of disability worldwide. Stroke patients are at risk of developing a wide range of medical complications during the acute phase and later during the post-stroke rehabilitation period. The complications can significantly hinder optimal recovery and are potentially life-threatening. There is a high incidence of complications, from 48% to 96%, among post-stroke patients during hospitalization in neurorehabilitation units (NU) (Davenport et al., 1996; Doshi et al., 2003; Dromerick & Reding, 1994; Kalra et al., 1995; Langhorne et al., 2000; Roth et al., 2001). The most common types of medical complications that occur in NU are: urinary tract infections, falls, pain, depression, venous thromboembolism, cardiac diseases, and other infections (Davenport et al., 1996; Domka et al., 2005; Doshi et al., 2003; Dromerick & Reding, 1994; Indredavik et al., 2008; Ingeman et al., 2011; Johnston et al., 1998; Kalra et al., 1995; Langhorne et al., 2000). Studies show that complications impede the rehabilitation process and are closely related to a poor functional outcome (Saxena et al., 2006). Patients in unstable condition are generally not able to implement rehabilitation programs of equal intensity as stable patients. Various studies have shown that approximately 7% to 19% of stroke patients undergoing rehabilitation need to be transferred to other units due to complications requiring specialist treatment (Dromerick & Reding, 1994; Kalra et al., 1995; Roth et al., 2001).

Complications occur more often in older patients, in those who were disabled before stroke, and in those with severe strokes (Davenport et al., 1996). The severity of stroke was strongly associated with a greater risk of the majority of complications (Davenport et al., 1996; Indredavik et al., 2008; Langhorne et al., 2000; Roth et al., 2001). Many of these problems are potentially preventable or treatable if recognized early enough. Stroke patients usually have comorbidities such as hypertension, diabetes, heart disease, or other ailments that increase the risk of complications during recovery. The effectiveness of organized stroke care, improvements in prevention, early recognition, and correct treatment of complications can considerably reduce mortality. Therefore, the neurologist’s and the neurorehabilitation physician’s roles are to manage the most common comorbidities in the acute phase of stroke and prevent complications related to the direct effect of the neurologic injury, unstable comorbidities, or medication side effects (Dohle & Reding, 2011).

Studies on the occurrence of post stroke complications have been published; however, there are several discrepancies between reports concerning the frequency and type of complications (Doshi, 2003; Dromerick & Reding, 1994; Hung, 2005; Kalra 1995; Langhorne, 2000; Roth, 2001). These differences might have origin in different sample size starting with 100 patients in one of the oldest report (Dromeric & Reding, 1994), through 245 (Kalra et al., 1995), 311 (Langhorne et al., 2000), 140 (Doshi et al., 2003), 346 (Hung et al., 2005), up to 1029 (Roth et al., 2001). Some discrepancy might be also caused by different methods of data collection (time from stroke onset to starting institutionalized rehabilitation) and differences in patients’ characteristics. The study of Kalra et al. included stroke patients treated either on a stroke rehabilitation unit or on general medical wards early (2 weeks) after stroke (Kalra et al., 1993, 1995). Medical complications occurred in 60% of patients and were more frequent in patients with severe strokes (97%). The rate of complications was similar in both settings. Aspiration and musculoskeletal pain were more often documented on the stroke unit, while urinary problems and infections were more common on general medical wards. In turn, Langhorne et al. recruited stroke patients admitted to 3 hospitals within 7 days of stroke onset, and their progress was examined on a weekly basis until discharge from hospital and afterwards at 6, 18, and 30 months after stroke. A total of 265 patients (85%) had at least 1 complication during their stay in hospital. The most common were: falls, infections (including urinary and respiratory tract), pain (including shoulder pain) and depression. Doshi et al. conducted retrospective case review for the six-month period. The patients were Chinese, Malay and Indian. The most common complication noted in this study was constipation, not reported in the previous studies. A retrospective analysis was also performed by Hung et al. who included patients whose stroke occurred within the prior 3 months. 44% of patients had at least one medical complication, including the most common – musculoskeletal pain, urinary tract infection and depression. Pneumonia was observed much rarely. The study by Roth et al. (2001) is the largest one with 1029 patients admitted to the specialty rehabilitation hospital, whose stroke occurred within prior 3 months. The race of the patients was diverse, including 57% of white. The most commonly reported complications in this study were urinary tract infections, joint or soft tissue pain and depression.

The aims of the present work were to determine the incidence of medical complications that occur during early inpatient rehabilitation and to assess the influence of medical complications on the rehabilitation outcome, measured with the Barthel Index (BI) at discharge.

2 Materials and methods

We analyzed the clinical data of all patients hospitalized in the Neurorehabilitation Unit (NU) 2nd Department of Neurology, Institute of Psychiatry and Neurology, Warsaw, Poland from 2006–2010. Stroke diagnosis was established in the acute stroke unit according to the World Health Organization (WHO) definition (Sacco, 2013), and was confirmed by computed tomography (CT) or magnetic resonance imaging (MRI). Qualification criteria for the NU were: time from stroke onset to admission to NU up to 100 days, acute neurological deficit resulting in a significant reduction of activity, stable general condition, relatively preserved ability to learn, toleration of sitting position with support, and consent for participation for in-hospital rehabilitation. All patients fulfilling the qualification criteria for the NU were admitted, without any age limitation according to Experts of Division for Cerebrovascular Diseases of PNS (2012). Every patient received post-stroke rehabilitation therapy appropriate for personal clinical needs during the early period of stroke recovery. The rehabilitation therapy was compliant to the recommendations of the National Health Fund. Each patient received multidisciplinary rehabilitation program – conducted for total of about 150 minutes rehabilitation therapy every 5 days a week. Every patient had both: individual (one-to-one) and group physiotherapy and also occupational therapy. Individual physiotherapy and occupational therapy lasted about 40 minutes each. Patients with aphasia had additionally 30 minutes of speech therapy 5 days a week. For subjects with impaired cognitive functions (eg. attention, memory, visuospatial or executive dysfunctions) neuropsychological therapy was additionally carried out – 30 minutes 5 days a week. On Saturdays every patient took part in a group physiotherapy for at least 40 minutes. The physical therapy was used in a total of 251 patients (24 %).

Data concerning stroke risk factors and clinical conditions were collected prospectively according to the protocol routinely used in the department (Czernuszenko & Czlonkowska, 2009). For all patients, the following data were obtained: demographic characteristics (age, gender), type of stroke (ischemic, hemorrhagic), comorbidities (hypertension, diabetes, heart disease, heart failure, atrial fibrillation, myocardial infarction, intermittent claudication, smoking, alcohol consumption), medication used during the patient’s stay in the rehabilitation ward (antidepressive drugs, neuroleptics, anticonvulsants, anti-parkinsonian agent, hypotensive drugs, diuretics, benzodiazepines and other sedatives, antiplatelet agents, heparin, warfarin/acenocoumarol). The study data were collected from patients as part of their routine admission/discharge medical assessments. All assessments and records were made by physicians taking care of the patients during hospitalization.

2.1 Clinical, functional, and outcome assessment

Neurological impairment was measured with the National Institute of Health Stroke Scale – NIHSS (Brott et al., 1989) at admission (NIHSS_0) and at discharge (NIHSS_1). The Rankin scale (RS) was used to assess pre-rehabilitation (RS_0) and post-rehabilitation (RS_1) dependence (Van Swieten et al., 1988; Wilson et al., 2002). The activities of daily living were assessed using the BI (Collin et al., 1988; Mahoney & Barthel, 1965) at admission (BI_0) and at the patient’s discharge from the rehabilitation ward (BI_1). Patients were categorized according to BI scores into four groups concerning the severity of disability: mild (>15), moderate (11–15), mild-severe (6–10), and severe (<6). Outcome was defined according to the BI score group. The duration of time between stroke onset and admission to the NU, length of stay (LOS) in the rehabilitation ward, and the place of discharge were documented.

2.2 Complications

Medical complications were defined as the appearance or evident exacerbation of already existing symptoms, measures, or laboratory findings during hospitalization in the NU. Major complications that might disturb the rehabilitation process were documented, including infections, falls, and post-stroke depression, with regard to all patients included in the study. The complications were defined according to criteria described previously (Czernuszenko & Czlonkowska, 2009; Davenport et al., 1996; Langhorne et al., 2000), as follows:

falls – sudden events regardless of cause resulting in uncontrolled position change or coming to a lower level, recorded by the staff members who witnessed them;

fractures – confirmed by X-ray examination;

urinary tract infections – clinical symptoms or positive urine culture;

lower respiratory tract infections – auscultatory respiratory crackles, cough, fever, dyspnea, radiographic evidence;

hypertension – new onset and blood pressure requiring modification of antihypertensive therapy;

diabetes – new onset and glycemic level requiring modification of diabetic treatment;

depression – low mood requiring pharmacological or psychiatric intervention;

neurological: recurrent stroke (CT, MRI), seizure – clinical diagnosis of seizure in a previously non-epileptic patient;

cardiac events – clinical diagnosis by ECG evidence, exacerbation of ischemic heart disease;

diarrhea – bowel movements >three times/day, loose stool consistency;

thromboembolism – deep vein thrombosis with pain, swelling, redness, warmness, positive for Homans’ sign, elevated D-dimer levels (>500 ng/ml), found by USG-Doppler of lower limb veins or pulmonary embolism with dyspnea, cough, elevated D-dimer levels (>500 ng/ml), found by CT of the chest;

shoulder pain – pain in the shoulder area requiring analgesia on two or more consecutive days;

other pain – joints and soft tissues, neuropathic pain;

other complications – any documented complication resulting in medical or surgical intervention (e.g. gastrointestinal hemorrhage, cholelithiasis, bedsore).

For these estimates, a particular complication was recorded only once for a patient. Data concerning common complications were obtained prospectively from medical documentation.

2.3 Statistical analysis

The results of the descriptive analysis are presented as means, standard deviations, medians, and ranges. The Wilcoxon Rank-Sum test was used for comparisons of two independent groups. Bonferroni correction was used for multiple comparisons. The prognostic value of BI_1 was determined using the initial state value BI_0 and risk factor levels. The effects of the risk factors on the value of Bl_1 were calculated using multidimensional generalized linear models. The most appropriate relation was obtained for selected factors with a logarithmic link function and normal distribution of random errors. Statistical tests were evaluated at the 0.05 level of significance. Statistical analyses were carried out with the SAS9.2 (USA, 2010).

3 Results

The registry included 1236 consecutive patients with a stroke diagnosis admitted to the hospital. Of these, 1075 early post-stroke (defined as an admission to the NU within 100 days from the stroke onset) patients were included in the study: 876 (81.5%) with ischemic and 199 (18.5%) with primary hemorrhagic strokes. There were 161 post-stroke patients not included in our study because they were admitted to the NU in late and chronic phases of stroke rehabilitation (the time from the stroke onset to admission to the NU was >100 days, most of them >6 months after the stroke onset). Both Bl_0 and BI_1 were recorded in 1029 of the 1075 patients.

3.1 Patients’ demographics and clinical characteristics

The patients’ demographics, clinical characteristics, and risk factors before admission to the NU are presented in Table 1.

3.2 Incidence and types of complications in studied groups of patients

Less than a quarter of patients had no complications during rehabilitation. The most common complications were: urinary tract infections (23.2%), depression (18.9%), falls (17.9%), unstable hypertension (17.6%), and shoulder pain (14.9%). Less frequent complications were: respiratory tract infections (8.8%), cardiac events (6.5%), diarrhea (6.1%), diabetes (5.8%), other pain (5.2%), deep vein thrombosis (1.5%), seizure (1.2%), recurrent stroke and pulmonary embolism (0.7%), and fractures (0.5%). Nineteen patients fell twice, five patients – three times. Nearly half of the patients (45.5%) had more than one complication (Table 2).

There was a negative correlation between the BI score at admission and the number of complications experienced in the NU and a positive correlation between the RS and NIHSS at admission and the number of complications experienced: r = –0.43, r = 0.39, r = 0.36 (p < 0.0001 for all correlations). Average numbers of days of hospitalization were as follows: 32.3 ± 14.2 days for patients without any complications, 31.4 ± 19.3 days for patients with one complication, 37.2 ± 20.6 days for patients with 2 complications, 42.8 ± 21.1 days for patients with 3 complications and 47.6 ± 20.9 days for more than 3 complications. It gives mean 5 days longer hospital stay per 1 complication more, starting from 2 complications. Statistically, the number of complications had a slight impact on the length of hospitalization, especially in patients with the most severe disability (BI < 6), (r = 0.17, p = 0.0013). There was also a weak positive correlation between age and the number of complications after a stroke (r = 0.1, p = 0.0013).

3.3 Comparison of clinical parameters according to BI groups

The distribution of the four disability levels assessed by BI score at admission to rehabilitation was as follows: 259 patients (25.2%) with mild disability (group 1), 212 patients (20.6%) with moderate disability (group 2), 221 patients (21.5%) with moderate to severe disability (group 3), and 337 patients (32.8%) with severe disability (group 4). In the group of patients with severe disability, 55.8% were women and the mean patient age was the highest of all groups, 66.1 years compared with 62.1 years in group 1 (p < 0.0001), 63.5 years in group 2 (p = 0.03), and 64.9 years in group 3 (p = 0.22). The mean time from stroke onset to admission to the NU was the longest in group 4, 33.6 days vs. 29.4 days in group 1 (p = 0.0018), 29.9 days in group 2 (p = 0.04), and 31.2 days in group 3 (p = 0.15). Similarly, the patients with severe disability stayed the longest in the NU: 43.7 days compared with 25.7 days in group 1 (p < 0.0001), 30.6 days in group 2 (p < 0.0001), and 40.1 days in group 3 (p = 0.06). Over 3-fold more patients with severe disability at admission were transferred to other wards compared with patients with mild disability at admission (17.2% vs. 5%, p < 0.0001). The most severe disabled group had 2.5-fold more complications than those with the lowest disability, p < 0.001 (Table 3).

3.4 Factors affecting stroke rehabilitation outcome

The analysis revealed that sex, admission neurological examination score (NIHSS_0), the occurrence of complications (hypertension, cardiac events, recurrent stroke, diarrhea, pulmonary embolism), length of stay, and time from stroke onset to rehabilitation were predictors affecting the BI score at patient discharge (Table 4). In addition, patients who stayed in the NU for 5-6 weeks had similar chances of getting the same score on the BI at discharge as those who stayed longer (7–16 weeks). This relationship was especially strong in patients with a BI score at admission of <13 (Fig. 1).

Our research also showed that the number of complications affects the final result and the functioning of a patient after discharge from the hospital. A lack of complications during hospitalization significantly increased the chances for improvement. This relation was particularly noticeable in patients with severe disability at admission. Severely impaired patients (low BI on admission) had on average of a 5 point lower BI at discharge than patients with the same level of initial impairment but presenting no complications. (Fig. 2). In patients with mild disability (BI_0 > 15), the number of complications did not affect functional outcome measured with the BI_1. Only 3.9% of patients with a low initial level of disability (BI_0 > 15) on admission experienced 3 or more complications comparing to 37% of severely impaired patients.

Patients with mild disability without complication did not differ significantly from patients with one, two, or three or more complications (p = 0.34, p = 0.40, p = 0.45, respectively). This suggests that complications are less important in terms of final functional outcome (BI_1) in groups with higher BI_0 scores.

4 Discussion

The current study was performed to determine the frequency and type of general complications in a representative group of post-stroke patients admitted to the NU within a limited time-window (100 days from stroke onset). Data were collected prospectively using the same methodology as well as rehabilitation and treatment criteria. Physiotherapy was performed by the same staff. The frequency of complications noted in our study (76.9%) is comparable with some previously made observations (72% and 75%) (Brola et al., 2008; Roth et al., 2001). Nearly half of the patients (45.5%) had more than one complication, and one of five patients had three or more complications. Therefore, more attention should be paid to the prevention, early diagnosis and treatment of complications. For example, as UTIs are the most frequent complications, the need for urinary catheters should be assessed daily and these should be discontinued as soon as possible to avoid infections. Special attention should be payed to signs of depression, unstable hypertension, falls and shoulder pain, especially because the majority of these complications are at least treatable if not preventable.

Our findings concerning occurrence of different complications confirm previous observations (Brola et al., 2008; Davenport et al., 1996; Dromerick et al., 1994; Langhorne et al., 2000; McLean, 2004) that post-stroke patients are at high risk for urinary tract infections, depression, falls, and pain. Our finding of low frequencies of recurrent stroke, post-stroke seizure, deep vein thrombosis, and pulmonary embolism in the present research are similar to those observed previously (Roth et al., 2001).

In the current study, complications occurred 2.5-fold more often in patients with severe disability (BI_0 < 6) compared with patients with mild disabilities (BI_0 > 15 points). The association between the number of complications and disability measured by BI has been previously described. Langhorne et al. (2000) found a gradually increasing risk of infection, bedsores, anxiety, depression, and falls in patients with more severe disability, measured with the Functional Independence Measure. In recent years, after the implementation of organized post-stroke care, there has been a significant increase in the number of post-stroke patients with severe conditions qualifying for rehabilitation in specialized units. According to the prognoses, this trend will continue due to demographic changes in developed countries – mainly aging. Therefore, specialists involved in the rehabilitation process should be well-educated in prevention, early recognition, and the treatment of complications.

The number of complications affects the final result and the functioning of patients after discharge from the hospital, according to the BI. A lack of complications during hospitalization significantly increases the chance for improvement. The patient’s condition at admission, expressed in each scale (BI, mRS, NIHSS), correlated with the number of complications; although, the Spearman correlation coefficient was very low (r = –0.43, 0.39, 0.36). However, the results of our study are in accordance with previous reports (Brola et al., 2008; Kalra et al., 1995; Roth et al., 2001). Our study is one of the few carried out in such a large population of stroke patients in the NU. In previous research, patient populations were much smaller, up to 350 patients (Domka et al., 2005; Doshi et al., 2003; Dromerick et al., 1994; Hung et al., 2005; Kalra et al., 1995; Kitisomprayoonkul et al., 2010; Langhorne et al., 2000; McLean, 2004). Only one study by Roth et al. (2001) was based on a similar number of patients. The study included 1029 patients vs. 1057 in our study. However, the time to admission to the NU was much shorter than in the current work with a mean of 17 days vs. 32 days. The complication rate (75%) was almost the same as in our study (76.9%). Patients studied by Roth were divided into four groups in terms of the neurological deficit measured with the NIHSS. The smallest group – 13% was the one with the largest deficit (NIHSS >16); whereas in our study, patients with a severe disability (BI < 6) constituted the largest group (33%). Moreover, the population studied by Roth et al. was different in terms of race, 57% of subjects were white; while, all patients in our study were white. Results of the previous studies suggest that ethnic differences may influence the severity, prognosis, and outcome of rehabilitation after stroke (Bhandari et al., 2005; Kurian & Cardarelli, 2007; Liebson, 2010; McGruder et al., 2004; Ottenbacher et al., 2008; Stansbury et al., 2005). Our study provided additional results supporting the existing data.

The severity of stroke is considered the strongest predictor of functional improvement (Abdul-Sattar & Godab, 2013). In the current study, patients with a NIHSS <6 points (mild stroke) were most likely to get a higher score on the BI at discharge from NU. The BI at discharge strongly depends on the neurological status at admission. Similarly, Shah et al. (1989) reported that disability at admission to the rehabilitation unit was a strong predictor of the BI result at discharge. One of the observations of our study was that patients with a 6 week length of stay in the NU had similar chances of getting the same activities of daily living score at discharge as those who stayed longer. Similar phenomenon has already been described by Kwakkel et al. in 2003. This group conducted the study on a homogenous group of 102 patients with first-ever severe middle carotid artery (MCA) stroke resulting in complete hemiplegia. All patients were severely disabled (BI ≤ 9 points). The study showed that 62% of patients were unsuccessful in achieving improvements in dexterity at 6 months, indicating that the prognosis for functional outcome in MCA stroke is poor. Authors suggested that optimal prediction of arm function outcome was limited to only 4 weeks after onset.

Our findings lead to the conclusion that after 6 weeks, the patient should be re-examined to determine new goals and possible effects of further rehabilitation or to discuss discharge. Therefore, we have to consider all possible circumstances that could impact outcome.

5 Conclusions

Rehabilitation is a long-term process requiring the cooperation of an interdisciplinary team and should begin as soon as possible after the onset of stroke and last as long as improvement continues. The knowledge of the prevalence of certain complications and their impact on the course of rehabilitation can help minimize their negative effects. The decision to discharge is based on the individual and a prolonged hospital stay should not be a routine practice.

Conflict of interest

The authors declare that they have no conflicts of interest.

Footnotes

Acknowledgments

Special thanks to Mrs. Anna Karwanska for her statistical support. Research subject implemented with CePT infrastructure financed by the European Regional Development Fund within the “Innovative economy for 2007–2013” operational program.

References

Abdul-Sattar

A. B

, & Godab

(2013). Predictors of functional outcome in Saudi Arabian patients with stroke after inpatient rehabilitation. Neuro Rehabilitation, 1(33), 209–216.

Bhandari

V. K

, Kushel

, Price

, & Schillinger

(2005). Racial disparities in outcomes of inpatient stroke rehabilitation. Arch Phys Med Rehabil, 86, 2081–2086.

Brola

, Fudala

, Przybylski

, & Czernicki

(2008). Prevention of late complications of stroke. Studia Medyczne, 9, 21–26 (in Polish).

Brott

T. G

, Adams

H. P

Jr , Olinger

C. P

, Marler

J. R

, Barsan

W. G

, Biller

, Spilker

, Holleran

, Eberle

, & Hertzberg

(1989). Measurements of acute cerebral infarction: A clinical examination scale. Stroke, 20, 864–870.

Collin

, Wade

D. T

, Davies

, & Horne

(1988). The Barthel ADL Index: A reliability study. Int Disabil Stud, 10, 61–63.

Czernuszenko

, & Czlonkowska

(2009). Risk factors for falls in stroke patients during inpatient rehabilitation. Clin Rehab, 23, 176–188.

Domka

, Myjkowska

, & Kwolek

(2005). Evaluation of the incidence of complications in patients undergoing rehabilitation after stroke. PJNNS, 39, 300–309 (in Polish)

Doshi

V. S

, Say

J. H

, Young

S. H

, & Doraisamy

(2003). Complications in stroke patients: A study carried out at The Rehabilitation Medicine Service, Changi General Hospital. Singapore Med J, 44, 643–652.

Davenport

R. J

, Dennis

M. S

, Wellwood

, & Warlow

(1996). Complications after acute stroke. Stroke, 27, 415–420.

10.

Dohle

C. I

, & Reding

M. J.

(2011). Management of medical complications. Neurorehabilitation, 17, 510–529.

11.

Dromerick

, & Reding

(1994). Medical and neurological complications during inpatient stroke rehabilitation. Stroke, 25, 358–361.

12.

Experts of Division for Cerebrovscular Diseases of PNS (2012). Proceeding in stroke. PJNNS, 46(supl 1), 87–93 (in Polish).

13.

Hung

J. W

, Tsay

T. H

, Chang

H. W

, Leong

C. P

, & Lau

Y. C.

(2005). Incidence and risk factors of medical complications during inpatient stroke rehabilitation. Chang Gung Med J, 28, 31–38.

14.

Indredavik

, Rohweder

, Naalsund

, & Lydersen

(2008). Medical complications in a comprehensive stroke unit and an early supported discharge service. Stroke, 39, 414–420.

15.

Ingeman

, Andersen

, Hundborg

H. H

, Svendsen

M. L

, & Johnsen

S. P.

(2011). In-hospital medical complications, length of stay, and mortality among stroke unit patients. Stroke, 42, 3214–3218.

16.

Johnston

K. C

, Li

J. Y

, Lyden

P. D

, Hanson

S. K

, Feasby

T. E

, Adams

R. J

, Faught

R. E

Jr , Haley

E. C

Jr , (1998). Medical and neurological complications of ischemic stroke: Experience from the RANNTAS trial. Stroke, 29, 447–453.

17.

Kalra

, Yu

, Dale

, & Crome

(1993). Improving stroke rehabilitation. A controlled study. Stroke, 24, 1462–1467.

18.

Kalra

, Yu

, Wilson

, & Roots

(1995). Medical complication during stroke rehabilitation. Stroke, 26, 990–994.

19.

Kurian

A. K

, & Cardarelli

K. M.

(2007). Racial and ethnic differences in cardiovascular disease risk factors: A systematic review. Ethn Dis, 17, 143–152.

20.

Kwakkel

, Kollen

B. J

, van der Grond

, & Prevo

A. J.

(2003). Probability of regaining dexterity in the flaccid upper limb: Impact of severity of paresis and time since onset in acute stroke. Stroke, 34, 2181–2186.

21.

Langhorne

, Stott

D. J

, Robertson

, MacDonald

, Jones

, McAlpine

, Dick

, Taylor

G. S

, & Murray

(2000). Medical complications after stroke: A multicenter study. Stroke, 31, 1223–1229.

22.

Liebson

P. R.

(2010). Cardiovascular disease in special populations III: Stroke. Prev Cardiol, 13, 1–7.

23.

Mahoney

F. I

, & Barthel

(1965). Functional evaluation: The Barthel Index. Maryland State Med J, 14, 56–61.

24.

McLean

D. E.

(2004). Medical complications experienced by a cohort of stroke survivors during inpatient, tertiary-level stroke rehabilitation. Arch Phys Med Rehabil, 85, 466–469.

25.

McGruder

H. F

, Malarcher

A. M

, Antoine

T. L

, Greenlund

K. J

, & Croft

J. B.

(2004). Racial and ethnic disparities in cardiovascular risk factors among stroke survivors. United States 1999-2001. Stroke, 35, 1557–1561.

26.

Ottenbacher

K. J

, Campbell

, Kuo

Y. F

, Deutsch

, Ostir

G. V

, & Granger

C. V.

(2008). Racial and ethnic differences in postacute rehabilitation outcomes after stroke in the United States. Stroke, 39, 1514–1509.

27.

Roth

E. J

, Lovell

, Harvey

R. L

, Hainemann

A. W

, Semik

, & Diaz

(2001). Incidence of and risk factors for medical complications during stroke rehabilitation. Stroke, 32, 523–529.

28.

Sacco

R. L

, Kasner

S. E

, Broderick

J. P

, Caplan

L. R

, Connors

J. J

, Culebras

, Elkind

M. S

, George

M. G

, Hamdan

A. D

, Higashida

R. T

, Hoh

B. L

, Janis

L. S

, Kase

C. S

, Kleindorfer

D. O

, Lee

J. M

, Moseley

M. E

, Peterson

E. D

, Turan

T. N

, Valderrama

A. L

, Vinters

H. V

American Heart Association Stroke Council, Council on Cardiovascular Surgery and Anesthesia; Council on Cardiovascular Radiology and Intervention; Council on Cardiovascular and Stroke Nursing; Council on Epidemiology and Prevention; Council on Peripheral Vascular Disease; Council on Nutrition, Physical Activity and Metabolism. (2013). An updated definition of stroke for the 21st century: A statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke, 44, 2064–2089.

29.

Saxena

S. K

, Ng

T. P

, Yong

, Fong

N. P

, & Gerald

(2006). Total direct cost, length of hospital stay, institutional discharges and their determinants from rehabilitation settings in stroke patients. Acta Neurol Scand, 114, 307–314.

30.

Shah

, Vanclay

, & Cooper

(1989). Predicting discharge status at commencement of stroke rehabilitation. Stroke, 20, 766–769.

31.

Stansbury

J. P

, Jia

, Williams

L. S

, Vogel

W. B

, & Duncan

P. W.

(2005). Ethnic disparities in stroke: Epidemiology, acute care, and postacute outcomes. Stroke, 36, 374–386.

32.

Van Swieten

J. C

, Koudstaal

P. J

, Visser

M. C

, Schouten

H. J

, & van Gijn

(1988). Interobserver agreement for the assessment of handicap in stroke patients. Stroke, 19, 604–607.

33.

Wilson

J. T

, Hareendran

, Grant

, Baird

, Schulz

U. G

, Muir

K. W

, & Bone

(2002). Improving the assessment of outcomes in stroke: Use of a structured interview to assign grades on the modified Rankin Scale. Stroke, 33, 2243–2246.